User: Rory Stark

gravatar for Rory Stark
Rory Stark1.8k
Reputation:
1,810
Status:
Trusted
Location:
CRUK, Cambridge, UK
Last seen:
5 days, 2 hours ago
Joined:
4 years, 4 months ago
Email:
r*********@cruk.cam.ac.uk

I am Principal Bioinformatics Analyst at the University of Cambridge, leading the Analysis Team in the Bioinformatics Core at Cancer Research UK's Cambridge Institute.

Posts by Rory Stark

<prev • 331 results • page 1 of 34 • next >
0
votes
3
answers
108
views
3
answers
Comment: C: DiffBind : More fragment sizes than libraries
... I've checked in a change so that if you set initString=NULL, it will not add the "_", so you can specify a full filepath in the "file" parameter. ...
written 5 days ago by Rory Stark1.8k
0
votes
1
answers
63
views
1
answers
Answer: A: DiffBind :Retrieve count value from Diffbind by dab.report
... The transform is a little different than rlog. The read counts are, as you saw above, normalized by the minimum library size. The concentrations are the log2() of the mean of these values. You can do a "pure" DESeq2 normalization by setting bFulLibrarySize=FALSE in the call to dba.analyze(). This w ...
written 5 days ago by Rory Stark1.8k
0
votes
1
answers
50
views
1
answers
Answer: A: Diffbind: retrieve raw peaks per sample
... For retrieving a specific peakset, you need to specify the peakset number if you want just the one: raw_peak <- dba.peakset(dba_chip_init,peaks=1,bRetrieve=T,DataType=DBA_DATA_GRANGES) The sampID parameter is inoperative in this case. -Rory   ...
written 5 days ago by Rory Stark1.8k
0
votes
3
answers
149
views
3
answers
Comment: C: DiffBind Batch Effect
... Having modelled the batch effect, the resulting statistics should be valid. In particular, I would expect that False Positive statistics to be valid. Yo may have more False Negatives formt he increased variance, but he sites that have low FDR should be good. -R ...
written 5 days ago by Rory Stark1.8k
0
votes
3
answers
108
views
3
answers
Comment: C: DiffBind : More fragment sizes than libraries
... See this discussion: Question: Binding site overlaps -R ...
written 8 days ago by Rory Stark1.8k
0
votes
3
answers
108
views
3
answers
Comment: C: DiffBind : More fragment sizes than libraries
... Not sure what you are trying to do with the second line? the value of 1 will be assigned to the second argument, mask, so this will return the number of peaks in the first sample. It is the same as saying: dba_first <- dba(dba_chip, mask=1) dba.overlap(dba_first,mode=DBA_OLAP_RATE) If you just ...
written 8 days ago by Rory Stark1.8k
0
votes
1
answers
249
views
1
answers
Comment: C: DiffBind BED output with score
... As far as standard ChIP-seq data goes, the protocol is not strand-specific -- there is no way to know where on the pulled-down DNA, or which strand, the protein of interest was binding. Peaks form peak callers generally do not have a strand specified, the entire genomic interval (both strands) compr ...
written 9 days ago by Rory Stark1.8k
0
votes
3
answers
149
views
3
answers
Comment: C: DiffBind Batch Effect
... The PCA and correlation plots may be misleading in this case. Modelling the batch effect in the GLM does not change the read counts in any way. So while the model used to identify differentially bound sites takes the batch effect into account, when you plot these sites using the count matrix, they w ...
written 10 days ago by Rory Stark1.8k
0
votes
8
answers
96
views
8
answers
Comment: C: got 0 significant differential bound (DB)
... You may want to plot your p-value distribution: > DBreport <- dba.report(myDBA, th=1) > hist(DBreport$"p-value") If the distribution is relatively flat, the FDR correction will be very harsh as that looks like an experiment where everything fit the null hypothesis. Here is a link to a go ...
written 10 days ago by Rory Stark1.8k
0
votes
3
answers
108
views
3
answers
Comment: C: DiffBind : More fragment sizes than libraries
... In this case your dba_chip_init objects looks like it is a DBA object (not a ChIPQCexperiment object). In that case, you can skip the first line and just set dba_chip_init$ChIPQCobj <- NULL and continue on your way: -R ...
written 10 days ago by Rory Stark1.8k

Latest awards to Rory Stark

Scholar 11 days ago, created an answer that has been accepted. For A: DiffBind "no lines available in input" error
Guru 6 weeks ago, received more than 100 upvotes.
Popular Question 6 weeks ago, created a question with more than 1,000 views. For a diffBind problem
Scholar 8 weeks ago, created an answer that has been accepted. For A: DiffBind "no lines available in input" error
Popular Question 9 weeks ago, created a question with more than 1,000 views. For a diffBind problem
Scholar 4 months ago, created an answer that has been accepted. For A: DiffBind "no lines available in input" error
Scholar 4 months ago, created an answer that has been accepted. For A: DiffBind "no lines available in input" error
Autobiographer 9 months ago, has more than 80 characters in the information field of the user's profile.
Scholar 14 months ago, created an answer that has been accepted. For A: DiffBind "no lines available in input" error
Scholar 14 months ago, created an answer that has been accepted. For A: What does dba.PCA in DiffBind exactly do?
Scholar 14 months ago, created an answer that has been accepted. For A: DiffBind matrix of values
Scholar 14 months ago, created an answer that has been accepted. For A: ChIPQC and/or Diffbind - retrieving normalized read counts for peaks from single
Teacher 15 months ago, created an answer with at least 3 up-votes. For A: Work with GRanges objects in DiffBind
Scholar 16 months ago, created an answer that has been accepted. For A: What does dba.PCA in DiffBind exactly do?
Scholar 18 months ago, created an answer that has been accepted. For A: What does dba.PCA in DiffBind exactly do?
Scholar 19 months ago, created an answer that has been accepted. For A: What does dba.PCA in DiffBind exactly do?
Scholar 20 months ago, created an answer that has been accepted. For A: Work with GRanges objects in DiffBind
Teacher 20 months ago, created an answer with at least 3 up-votes. For A: diffBind correlation plot
Teacher 20 months ago, created an answer with at least 3 up-votes. For A: Work with GRanges objects in DiffBind
Scholar 20 months ago, created an answer that has been accepted. For A: diffBind correlation plot
Scholar 20 months ago, created an answer that has been accepted. For A: Clarification about the functionality of dba.count and dba.analyze
Scholar 20 months ago, created an answer that has been accepted. For A: subset of samples from dba object in DiffBind
Scholar 23 months ago, created an answer that has been accepted. For A: problem with characters in chromosome names in DiffBind dba
Teacher 23 months ago, created an answer with at least 3 up-votes. For A: diffBind correlation plot
Scholar 2.1 years ago, created an answer that has been accepted. For A: diffBind correlation plot

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 149 users visited in the last hour